Guide for the Preservation of Highway Tunnel Systems (2015)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

18 C H A P T E R 5 5.1 Evaluating Alternative Preservation Actions To evaluate the effectiveness of a proposed tunnel preservation action, several factors must be considered. First, what effect does the improvement have on achieving the agency’s overall goals and objectives (i.e., meeting their LOSs)? Second, how cost-effective is the improvement (e.g., does it reduce maintenance or energy costs, what is the ultimate cost per user, and what effect does it have on the remaining life of the asset)? And finally, what are the associated risks if the improvement is delayed and how urgently is the improvement needed? While each of these considerations plays into the final prioritization made by tunnel owners, the final prioritization is often fairly subjective since the improvements vary significantly from a structural improvement to a tunnel system improvement. Furthermore, many agency bridge/ tunnel departments are led by structural engineers who understand the needs of bridges and the tunnel structure itself, but tunnels are complex facilities that involve multiple mechanical and electrical systems that must also be considered. Evaluating preservation actions to determine priorities can be a daunting task when many of the improvements needed are in different sys- tems. This chapter presents a strategy for measuring the effectiveness of preservation actions to facilitate comparisons of different improvements. 5.1.1 Risk Assessment, Management, and Mitigation Strategies Tunnel owners regularly assess risk when evaluating tunnel preservation actions, whether making long-term decisions on equipment needs or structural improvements. Risk miti- gation often has a direct impact on the safe operation of the tunnel, but risk must also be considered in terms of operational reliability, preservation, and the other agency goals. For example, the longer a piece of equipment goes without needed repairs due to lack of funds for replacement or rehabilitation, the greater its chance for failure. Its failure could result in an unsafe condition for the traveling public or require closure of the tunnel until repairs are made, or additional costs in maintenance of the equipment could result if it is left in its present condition. Similarly, the longer water leakage occurs within a tunnel, the greater the likelihood of further structural deterioration and the possibility of unsafe conditions for the traveling public. Without remediation, water infiltration may lead to extensive costs to repair deterioration of the structure and appurtenances and to eliminate the water problem. The 2011 AASHTO Transportation Asset Management Guide provides guidance for transportation agencies on this subject, recommending that they consider performing risk assessment and risk management as well as develop mitigation strategies as part of the routine business functions of the agency. Measuring Effectiveness of Preservation Actions

Measuring Effectiveness of Preservation Actions 19 The 2011 AASHTO Transportation Asset Management Guide provides an approach for risk assessment where the following are evaluated: • Likelihood of an extreme event, such as a flood, earthquake, asset failure, or other risk driver, expressed as a probability, or range of probability, of an event. • Consequences to the asset, a categorization of the damage or loss of function of the asset, and conditions on occurrence of an event. • Effect on mission, life, property, and the environment, a categorization of the effect on the agency, the public, users, and non-users, of the asset damage or loss of function caused by the extreme event.(5) A typical risk management framework was presented in the 2011 AASHTO Transportation Asset Management Guide and is shown in Figure 5-1. Owners should consider likelihood, consequences, and impact when evaluating risk. The level of risk can be categorized as low, moderate, high, and extreme, as shown in Figure 5-2 and as defined in the 2011 AASHTO Transportation Asset Management Guide. The asset management approach presented in this chapter suggests that owners evaluate risk and risk mitigation strategies for each preservation action as part of developing the various scores described in this chapter. These strategies should be promulgated by the AAMT and can serve as supporting data explaining why preservation actions are necessary when communicat- ing the needs to taxpayers or public officials. 5.2 Agency Levels of Service and Performance Goals As discussed in Chapter 3, each agency will have its own set of goals or its own specific LOSs. Preservation actions that focus on these LOSs may be given higher priority than those that do not because they assist in meeting the overall goals. In addition, certain LOSs will be more important Figure 5-1. Typical risk management framework. Figure 5-2. Risk likelihood and consequence categories. Consequences Likelihood Insignificant Minor Significant Major Catastrophic Very rare Low Low Low Moderate High Rare Low Low Moderate High High Seldom Low Moderate Moderate High Extreme Common Moderate Moderate High Extreme Extreme Frequent Moderate High High Extreme Extreme

20 Guide for the Preservation of Highway Tunnel Systems within the agency than others. For example, safety and preservation are two goals that typically rank higher than other LOSs for most highway agencies. One way to quantify how well a preserva- tion action improves the service level is to simply provide a score based on expert judgment or a qualitative assessment of its impact on performance. The LOS score would be one factor contrib- uting to the overall effectiveness of the preservation action. The examples that follow explain the process used to develop an LOS score for each preservation action. 5.2.1 Weighting the Levels of Service The AAMT will set relative weights for each LOS selected as part of determining an LOS score. Section 3.2 identified and described six general LOS standards that many transportation agen- cies are using as part of an asset management process. The tunnel owner may decide that all six are valid for its analysis, may select a reduced number, or may add others as it deems necessary. When all LOS standards are identified, the tunnel owner should assign an importance percentage to each LOS such that the total percentage for all LOS standards is 100%. These percentages serve as the weights. For setting percentages, the AAMT will rank in general order of importance from highest to lowest among the LOS standards for the tunnel at the time of the ranking. Agency X’s AAMT ranked the order of importance as: safety, reliability, preservation, quality of service, security, and environment. The ranking should reflect the long-term strategic directions of the agency. Factors influencing the establishment of this order could be as follows: • Safety was at the top of the AAMT’s evaluation when they considered the following factors: A number of accidents have occurred at a few of the agency’s tunnels; the seasonal snow accumulations cause safety concerns at those tunnel entrances in the colder climates; current pavement surface conditions, such as rutting, could cause safety concerns within the travel lanes; and visibility affected by contrasting light levels at all tunnel entrances in the daytime could be improved for better eye adjustment when entering the tunnels. • Reliability was deemed the second most critical factor in weighting as part of the asset manage- ment approach. This resulted from a series of unexpected events causing tunnel closure for a significant period of time. The agency’s tunnels carry high traffic volumes, and tunnel closure will affect the public due to the length of detours that may be needed or the congestion on other highways where traffic is routed. • Preservation was ranked third because maintaining a state of good repair implies that pres- ervation actions need to occur. Fans nearing the end of their useful life require extensive maintenance/repairs to keep the ventilation system operational, and water infiltration within the tunnel is causing the tiled surface to delaminate and spall, exposing the structural reinforcing in many areas. • Quality of service was ranked fourth because customers were complaining about the light levels present when they entered the tunnels and infiltrating groundwater dripping onto their vehicles when traveling through them. In addition, ride quality was reduced due to poor roadway pave- ment conditions, and tunnel aesthetics degraded as tunnel walls were dirty, stained, and needed to be cleaned to improve reflectivity and overall appearance. • Security was ranked fifth by the AAMT because security measures are already scheduled for implementation due to a security breach that recently occurred in a nearby facility. The security systems currently installed are obsolete and in need of replacement. In addition, the agency has received increasing numbers of special requests for moving hazardous/explosive materials through the tunnel during non–rush hours. While security is important, scheduled improve- ments result in security being less important than the other LOS standards. • Environment was ranked sixth since there have been few, if any, threats to the environment from the agency’s tunnels. Possible environmental concerns might include the draining of

Measuring Effectiveness of Preservation Actions 21 spills within the tunnel to local streams, exhausting of tunnel emissions in an urban area with air-rights structures, or disposal of tunnel lights containing hazardous materials; however, the agency deemed the likelihood of these kinds of issues in its tunnels to be low. This iterative process for consideration of ranking should continue within the AAMT until all members are in agreement with the ranking of LOS elements in order of importance. While developing the LOSs for its six tunnels, the AAMT agreed that improvements having a very high impact on safety should have priority above all others. Therefore, the agency has chosen to make repairs and improvements that “greatly” or “significantly” affect safety prior to other repairs and improvements. The remaining levels of service were assigned weights to total 100%, with reliability and preservation ranking second and third, respectively, after safety. The weights were assigned by the AAMT after ranking agency goals and are as shown in Table 5-1. When resulting LOS weights are less than or equal to 5%, the agency should evaluate whether to include those LOSs. In the case of this example, the agency should evaluate whether to include LOSs for security and environment. 5.2.2 Assessing the Impact of Preservation Actions on LOS Each preservation action will affect the agency’s ability to achieve the selected LOS, and this impact must be captured in the prioritization process. To weigh the relative impact of preserva- tion actions, each improvement must be evaluated with respect to each LOS. To evaluate how well an action addresses each LOS, a rating of 1 to 5 might be used, where a 5 indicates that the preservation action will greatly improve the performance associated with the LOS, and a 1 indi- cates that the preservation action will have very little impact. For each LOS, Table 5-2 defines the intent of the ratings 1 through 5. A rating of 0 would be used if the LOS was not applicable for that preservation action. The ratings are based on the subjective judgment of the tunnel owners’ trained inspectors and tunnel maintenance personnel. Each rating is intended to capture the degree to which the improvement contributes to the overall agency goals and objectives. For each of the agency’s selected preservation actions (see Section 4.3), LOS ratings are assigned as shown in Table 5-3. The rationale for selection of the LOS ratings for each of the preservation actions is provided in the following. • Ventilation upgrade to meet NFPA 502: Consulting Table 5-3, upgrading the existing aging ventilation greatly affects safety and preservation, but has very little impact on reliability (keeping the tunnel open to traffic), security, and quality of service. Appropriately, safety and preservation are assigned a rating of 5, and reliability, security, and quality of service are assigned a rating of 1. No environmental impact is anticipated from this improvement. • Install new light-emitting diode (LED) lights: The lights in Tunnel 1 are corroded and near the end of their service life; if not replaced soon, they run the risk of falling off the walls. There- fore, installing new light fixtures will significantly affect safety, and improved light levels will improve quality of service; these levels of service are assigned a rating of 4. After an analysis of alternative lighting, it was decided that LED fixtures should be installed. Removing the existing lights and installing the new ones greatly affects preservation, and this LOS is assigned a rating of 5. LED lights are much more energy efficient than the existing lights, and therefore the score for LOS Reliability Safety Security Preservation Quality of Service Environment Weight (%) 20 40 5 18 15 2 Table 5-1. Level-of-service weights.

22 Guide for the Preservation of Highway Tunnel Systems LOS Reliability Safety Security Preservation Quality ofService Environment Weights 20% 40% 5% 18% 15% 2% Preservation Action Tunnel # Ventilation upgrade to meet NFPA 502 1 1 5 1 5 1 N/A Install new LED lights 1 3 4 2 5 4 5 CO system – repair to operating condition 2 2 5 N/A 4 N/A 2 Repair active leak in tunnel 4 4 5 N/A 5 5 N/A Remove existing concrete tunnel ceiling 6 4 5 N/A 4 5 N/A Install flood gates 6 4 4 N/A 5 N/A N/A Table 5-3. Level-of-service ratings for selected preservation actions. LOS Ratings Reliability Safety Security Preservation Quality ofService Environment 1 The improvement will have very little… …impact on the ability to keep the tunnel open and operational. …impact on safety of workers or the traveling public. …impact on the vulnerability to technological or natural hazards. …effect on the remaining life of the asset. …effect on the experience for the driving public. …impact on the environment. 2 The improvement will somewhat… …affect the ability to keep the tunnel open and operational. …affect safety of workers or the traveling public. …affect the vulnerability to technological or natural hazards. …increase the remaining life of the asset. …improve the experience for the driving public. …affect the potential for environmental impacts. 3 The improvement will moderately… …improve the ability to keep the tunnel open and operational. …improve safety of workers or the traveling public. …reduce the vulnerability to technological or natural hazards. …increase the remaining life of the asset. …improve the experience for the driving public. …reduce the impacts or potential for environmental impacts. 4 The improvement will significantly… …improve the ability to keep the tunnel open and operational. …improve safety of workers or the traveling public. …reduce the vulnerability to technological or natural hazards. …increase the remaining life of the asset. …improve the experience for the driving public. …reduce the impacts or potential for environmental impacts. 5 The improvement will greatly… …improve the ability to keep the tunnel open and operational. …improve safety of workers or the traveling public. …reduce the vulnerability to technological or natural hazards. …increase the remaining life of the asset. …improve the experience for the driving public. …reduce the impacts or potential for environmental impacts. Table 5-2. Level-of-service ratings.

Measuring Effectiveness of Preservation Actions 23 environment is a 5. Although the current system provides the needed lighting levels to operate the tunnel, a failure could require tunnel closure, so reliability is assigned a 3. Due to the rural location of Tunnel 1, the new lights have little impact on security. • CO system—repair to operating condition: The existing CO system needs to be repaired to make it operational. Repairing the system greatly affects safety, and therefore this LOS receives the highest possible rating. Additionally, the service life of the asset may be extended after the repairs, so it has a significant impact on the preservation LOS. It has little impact on the ability to keep the tunnel open and operational, and has little impact on the environment. • Repair active leak in tunnel: There is a significant leak in Tunnel 4’s arch, which is causing deterioration of the structure and development of icicles in the winter months. Repairing the leak is needed to eliminate the cause of deterioration and to increase the service life of the structure; thus, preservation is rated a 5. Elimination of the infiltration will greatly increase the safety of the traveling public, since the current problem includes the possibility of falling icicles or slippery pavement. It will also eliminate the closures needed to remove icicles or delaminated concrete. While this preservation action may reduce the number of customer complaints about water dripping on their vehicles, repairing the leak has no effect on security levels or the environment. • Remove existing concrete tunnel ceiling: Tunnel 6 is located in an urban area and has signifi- cant traffic during peak hours. The tunnel has minimal horizontal and vertical clearances, and traffic typically slows significantly through the tunnel as a result. The ceiling has deteriorated through the years, and portions have been replaced at various times. Removing the existing tunnel ceiling will greatly affect the quality of service for motorists by improving the sense of openness and improving clearance and is anticipated to have a significant impact on the flow of traffic (reliability) through the tunnel. By eliminating the ceiling, safety is also greatly improved. Although the existing ceiling was recently repaired, the possibility of future spalling of the concrete will be eliminated with the ceiling’s removal. Removing the ceiling significantly affects the overall preservation of the tunnel by eliminating any future tunnel issues related to the existing deteriorated ceiling. The security and environment LOSs are not affected by this improvement. • Install flood gates: Installing flood gates has the potential to greatly affect the life of the tunnel and significantly improve safety. During a flood event, the gates will protect the tunnel structure and systems from damage and workers from dangerous conditions. The gates will allow the tunnel to be opened immediately after flood events since flooding of the tunnel and damage to its systems will be prevented. Therefore, it has a significant effect on reliability. No impact is expected on security, quality of service, or the environment. 5.2.3 Calculating the LOS Score The overall LOS score for each preservation action reflects the impact of the improvement on the agency LOS goals. The initial step is to identify the LOSs and their relative importance as defined in Section 5.2.1. Weights should be applied to each LOS, totaling 100%, based on the overall evaluation of their entire transportation network goals. This exercise may require con- siderable discussion among AAMT members until they arrive at percentages agreed upon by all members. The next step involves rating each preservation action’s impact on the performance associated with each rating as presented in Table 5-3. The aggregate LOS score can then be calculated using a weighted sum of the individual scores. The LOS score provides a measure of how well the preservation action would improve or help to achieve the agency’s LOS. The LOS score is calculated as follows: LOS 5 Equation 5-1W R W S W S W P W Q W ER S a S e P Q Ea e( )= ∗ + ∗ + ∗ + ∗ + ∗ + ∗

24 Guide for the Preservation of Highway Tunnel Systems where LOS = agency level-of-service score, R = reliability rating, Sa = safety rating, Se = security rating, P = preservation rating, Q = quality of service rating, E = environment rating, and WR, WSa, WSe, WP, WQ, WE = weights for reliability, safety, security, preservation, quality of ser- vice, and environment scores, where (WR + WSa + WSe + WP + WQ + WE) = 100. As discussed in Section 5.2.1, the AAMT decided to focus its priorities on actions that signifi- cantly and greatly affect safety (LOS ratings of 4 and 5). Table 5-4 shows a summary of ratings and final LOS scores for a selection of actions that have the desired impact on safety. Due to the high weighting of safety, the final LOS scores are, at a minimum, moderately high. Reliability and preservation are ranked second and third and are weighted similarly. Therefore, actions that affect safety, reliability, and preservation produce the highest LOS scores. Likewise, actions that have the least impact on safety, reliability, and preservation produce the lowest LOS scores. LOS score evaluations for all of Agency X’s preservation actions are included in Appendix D. 5.2.4 Calibrating the LOS Score There is a significant degree of subjectivity in the assignment of individual weights and the LOS ratings that are the basis of the LOS score. The weights assigned to each LOS will significantly Table 5-4. Preservation action ratings and LOS score. LOS Reliability Safety Security Preservation Quality of Service Environment LOS Score (Eq. 5-1)Weights 20% 40% 5% 18% 15% 2% Preservation Action Tunnel # Ventilation upgrade to meet NFPA 502 1 1 5 1 5 1 N/A 66.0 Install new LED lights 1 3 4 2 5 4 5 78.0 CO system – repair to operating condition 2 2 5 N/A 4 N/A 2 63.2 Repair active leak in tunnel 4 4 5 N/A 5 5 N/A 89.0 Remove existing concrete tunnel ceiling 6 4 5 N/A 4 5 N/A 85.4 Install flood gates 6 4 4 N/A 5 N/A N/A 66.0 Note: A rating of 0 is used when the LOS is not applicable for that preservation action.

Measuring Effectiveness of Preservation Actions 25 affect the resulting LOS score and, therefore, should be established by the AAMT and used consistently. As indicated in Section 5.1, the AAMT should follow a process where the LOSs are prioritized, and then individual weights are applied to each. An adjustment in the weights by a few percentage points up or down will result in similar variability in the resulting scores because an ultimate increase in one LOS weight requires a decrease in another to maintain a sum of 100. Similarly, there is a large degree of subjectivity in the assignment of LOS ratings for each pres- ervation action. It is recommended that the ratings be assigned by a technical team having specific knowledge of each of the improvements so that each preservation action can be accurately assessed for its impact on the LOS. The scoring from 1 to 5 was selected to limit the possible variations yet capture the variability. A smaller range of scores would have a larger impact on variability of the overall score, and a larger range of scores would make it difficult to differentiate between scores. It is likely that a preservation action considered to improve the LOS would be assigned a 4 or 5 if it were deemed to have a major impact, and a 1 or 2 if it would have little impact. Therefore, the variability in scoring, if the technical team has adequate knowledge of the improvement, should only result in a one-point difference in ratings. Since the highest weight possible for an LOS is 100%, the largest variation in LOS score for a one-point difference would be 20%. Since multiple LOSs will usually be assigned with varying weights, this helps to aggregate the scores. A one-point difference in LOS rating will therefore not usually result in a significant difference. In the example noted in Table 5-4, safety is weighted the highest (40%). For the ventilation improvement noted, a reduction in the safety rating from 5 to 4 would result in an LOS score of 58.0, a 12% reduction from the score of 66.0. Obviously, the greater the change in the product of LOS rating times LOS weight, the greater the variation in the overall score. 5.3 Cost-Effectiveness Cost plays a significant role in the comparison of tunnel maintenance and repair actions and in prioritizing actions for implementation. Tunnel owners will often implement improvements that are the least expensive because they fit within the overall budget and numerous improve- ments can be completed versus a few larger, more costly improvements. Sometimes contracting restrictions allow work within a specific dollar limit to proceed more expeditiously than more costly improvements. To compare alternatives using cost, life-cycle costing is typically used since it allows initial costs to be tempered by potential savings in maintenance and energy over the life of the asset. Tunnel assets, however, have varying service lives. Therefore, a means to compare life-cycle costs (LCCs) is to calculate the total LCC over a common period of time, such as 50 years. Alternatively, the LCC could be annualized over the service life to compare the costs per year. Another factor that plays into the decision of preservation action implementation is the num- ber of users that will be affected. A given improvement could have a high initial cost, but its impact would be felt by a large number of users of that particular tunnel. In this context, it is necessary to consider costs in terms of cost per user. Accordingly, improvements that have lower costs per user should receive higher priority. The ADT is used in the calculation of cost per user. This section provides a means of calculating a cost-effectiveness score to facilitate comparison of alternatives. 5.3.1 Life-Cycle Cost Life-cycle costing is a critical factor in the planning for future upgrades and repairs. LCCs are used to make decisions based on the life of an asset, incorporating possible cost savings or other costs that occur at a future date during the asset’s service life. It is beneficial to use LCC analysis

26 Guide for the Preservation of Highway Tunnel Systems when evaluating alternative approaches, whether alternative equipment or varying options for rehabilitation, to ensure the greatest cost efficiency over the life of the tunnel. This process involves evaluating the alternatives over a given duration or economic life to determine specific costs involved for each option and then equating them through a series of mathematical formulas that enable the costs of each option to be compared at a common point in time. A good reference on life-cycle costing analysis is FHWA’s Life-Cycle Cost Analysis Primer.(14) The LCC of a given alternative includes all associated costs over the expected life of the option. In general, these costs may include: • Initial costs: Project cost, including equipment and material costs, construction costs, and engineering or design costs. • Operating/energy costs: Annualized cost to operate (e.g., cost of electricity to run mechanical equipment, reduction in staff costs due to automation). These could be extra expenses or savings as compared to other alternatives such as the no-build alternative. • Maintenance costs: Annualized cost for maintenance. • Rehabilitation costs: Future expense for known procedures at a specified time (e.g., replacement of lamps, drivers/ballasts in lighting). • User costs: Costs associated with impact on the functioning of the tunnel (e.g., tunnel may need to be shut down for repair; therefore, impact to traffic can be shown by applying an annualized cost to each hour the tunnel is closed). • Salvage value: Sale value of equipment at the end of its service life (e.g., a mechanical fan may be of some value to others even after it has served its purpose in the tunnel). The data used for life-cycle cost calculations should be carefully considered. There is uncertainty when predicting the expected life and associated costs for an asset because there are many factors that affect asset function and costs. However, if quality information is used to evaluate each preservation action, then there is similar uncertainty among all actions allowing for an acceptable comparison of actions based on life-cycle costs. Evaluating LCC for different systems and tunnel elements with a wide range of service lives requires a common baseline for comparison. Calculating the present value (PV) of different pre- ventive actions or alternatives allows owners to account for the future value of their investments and associated maintenance and energy costs or savings. If all preventive actions were on the same time scale (i.e., all actions extended the remaining life by 20 years), then all could be compared using only the calculated present value. However, when actions extend the remaining life by different values, an annualized life-cycle cost is a more appropriate comparison measure. Two methods for performing a LCC analysis, the present worth method and the annualized method,(15) are outlined in the following. 5.3.1.1 Present Worth Method As the name implies, this method attempts to bring all of the present and future costs of a given option to present-day values. This process should be completed for each major repair/rehabilitation, and subsequently the present worth costs for each could be compared. Determining the present worth of a future expense is done by taking into account the time value of the money and, there- fore, discounting the amount by a predetermined rate (discount rate) over the period between the future expense and the present time. The present worth of the future expense is also the amount that could be invested today with reinvested interest over the duration to equal the amount of the future expense. An example of a future expense might be the cost to rehabilitate a piece of equip- ment that is being evaluated for replacement. The general form of the equation for determining the present worth of a future expense is: 1 1 Equation 5-2P F i n( )= +    

Measuring Effectiveness of Preservation Actions 27 where P = present worth, F = future one-time expense, n = number of years, and i = discount rate. Future expenses can also be uniform, in that the same expense occurs at the end of each year. An example of this would be the annualized maintenance costs described previously. The general form of the equation for determining the present worth of an end-of-year expense is: 1 1 1 Equation 5-3P A i i i n n ( ) ( )= + −  +    where P = present worth, A = end-of-year payments, n = number of years, and i = discount rate. 5.3.1.2 Annualized Method The annualized method is used to transform present and future costs into a uniform annual expense. This annual expense can be compared to the annual expenses of the other repair/ rehabilitation alternatives to determine which is most cost-effective. Converting all future expenses into a present value as before and then using Equation 5-4 to convert that value into an annual expense will provide a uniform annual cost. 1 1 1 Equation 5-4A P i i i n n ( ) ( )= +  + −    where P = present worth, A = end-of-year payments, n = number of years, and i = discount rate. The cash-flow diagrams in Figure 5-3 illustrate the annualized method. The annualized method allows for a simplified comparison of two preservation actions with different expected service lives (n, number of years) without having to use the same time scale for each preservation action. The cash-flow diagrams use the year-end convention, where it is assumed that all cash flows take place at the end of the year in which they occur (A, end-of-year payments). Once the equivalent annualized payment (Aeq) is determined for each action, the costs may be compared since consideration of the time scale is no longer required. Figure 5-3(a) displays the initial cash flow for a selected preservation action, including the capi- tal cost to implement the action, the annual change in costs per year, and a future one-time cost. The future one-time cost is moved to the present in Figure 5-3(b) using Equation 5-2 (P given F). In Figure 5-3(c), the annual change in costs is also moved to the present (P given A) with Equa- tion 5-3. Finally, in Figure 5-3(d), the present value of all of the cash flows is transformed into an equivalent annual cash flow (A given P) using Equation 5-4. These methods can also be performed using factors available from standard economic tables that are based on the discount rate and the economic life under consideration. These

28 Guide for the Preservation of Highway Tunnel Systems factors are also unique to the desired result. The procedure for using standard economic tables is as follows: • Determine the discount rate (i) and economic life (n) to be used for the analysis. It is impor- tant to choose an economic life that is equal for the given alternatives if the present worth method is to be used. Otherwise, the annualized method must be used. • Develop a cash-flow diagram for each option that shows all relevant costs described previously on a timeline of years in the economic life. • Take individual costs, whether uniform or one-time, and insert them in the proper formula given previously along with the factor from the appropriate economic table. – (P/F, i%, n)—or present worth (P) given future expense (F) at discount rate (i) for number of years (n). – (P/A, i%, n)—or present worth (P) given end-of-year payments (A) at discount rate (i) for number of years (n). – (A/P, i%, n)—or end-of-year payments (A) given present worth (P) at discount rate (i) for number of years (n). Time (n, years) Capital Cost Annual Change in Costs (A, occur each year) (a) Initial Cash Flow Time (n, years) Capital Cost + Present Value (PV) of F (b) Present Worth Cash Flow – Step 1 (Equation 5-2) Time (n, years) (d) Annualized Method Cash Flow (Equation 5-4) Equivalent Annual Cash Flows (Aeq, occur each year) Future One-Time Cost (F) Annual Change in Costs (A, occur each year) Time (n, years) Capital Cost + PV of F + PV of A (c) Present Worth Cash Flow – Step 2 (Equation 5-3) Figure 5-3. Annualized method.

Measuring Effectiveness of Preservation Actions 29 5.3.1.3 Discount Rate Caution should be taken when determining the appropriate discount rate. Because of the power of compounded interest, a difference in discount rate can actually change the final out- come of the analysis if the repair/rehabilitation options being considered have different arrange- ments of uniform and one-time costs. According to the FHWA draft report, A Discussion of Discount Rates for Economic Analysis of Pavements: The discount rate can affect the outcome of a life-cycle cost analysis in that certain alternatives may be favored by higher or lower discount rates. High discount rates favor alternatives that stretch out costs over a period of time, since the future costs are discounted in relation to the initial cost. A low discount rate favors high initial cost alternatives since future costs are added in at almost face value. In the case of a discount rate equal to 0, all costs are treated equally regardless of when they occur. Where alternative strategies have similar maintenance, rehabilitation, and operating costs, the discount rate will have a minor effect on the analysis and initial costs will have a larger effect.(16) For the purposes of this example, the discount rate was assumed to be 3%, reflecting the antici- pated average time value of money used for the analysis. The tunnel owner may elect to use the current national discount rate used by its agency for its calculations. 5.3.1.4 Life-Cycle Cost Example For Agency X’s preservation actions, the various cost elements making up each improvement must first be estimated, including the annual costs (potential savings) that apply over the service life of the improved asset. Table 5-5 shows the LCC analysis for the six preservation actions; the present worth of each action is calculated as an intermediate step for computing the annualized cost in the next section. Pr es er va tio n A ct io n Tu nn el # C ap ita l C os t ( $) A ge nc y O ve rs ig ht C os t ( $) A nn ua l E ne rg y C os t Sa vi ng s ( $) A nn ua l C ha ng e i n M ai nt en an ce C os t ( $) To ta l A nn ua l C ha ng e i n C os ts ($ ) Se rv ic e L ife o f t he A ss et A fte r Im pr ov em en t PV o f L C C ($ )** Ventilation upgrade to meet NFPA 502 1 5,700,000 570,000 –150,000 –2,500 –152,500 25 3,614,495 Install new LED lights 1 3,400,000 136,000 –66,000 –5,000 –71,000 20 2,479,699 CO system–repair to operating condition 2 32,000 3,200 0 0 0 20 35,200 Repair active leak in tunnel 4 10,000 1,000 0 0 0 20 11,000 Remove existing concrete tunnel ceiling 6 8,000,000 800,000 0 –20,000* –20,000* 50 8,285,405 Install flood gates 6 8,000,000 320,000 0 0 0 100 8,320,000 *Negative net change in annual cost denotes a cost savings. **PV of LCC is calculated based on the service life of the asset after the improvement is implemented. Table 5-5. Changes in annual costs and present value of LCC.

30 Guide for the Preservation of Highway Tunnel Systems Pr es er va tio n A ct io n Tu nn el # C ap ita l C os t ( $) A ge nc y O ve rs ig ht C os t ( $) A nn ua l C ha ng e i n C os ts ($ ) PV o f L C C ($ ) R em ai ni ng L ife D ue to P A A D T (x 10 00 ) A LC C ($ ) (E q. 5-5 ) A nn ua l C os t p er D ai ly V eh ic le ($ ) C E Sc or e ( Eq . 5 -6 ) Ventilation upgrade to meet NFPA 502 1 5,700,000 570,000 –152,500 3,614,495 25 40 207,573 5.19 1.9 Install new LED lights 1 3,400,000 136,000 –71,000 2,479,699 20 40 166,675 4.17 2.4 CO system – repair to operating condition 2 32,000 3,200 0 35,200 20 100 2,366 0.02 100.0 Repair active leak in tunnel 4 10,000 1,000 0 11,000 20 19 739 0.04 100.0 Remove existing concrete tunnel ceiling 6 8,000,000 800,000 –20,000 8,285,405 50 75 322,016 4.29 2.3 Install flood gates 6 8,000,000 320,000 0 8,320,000 100 75 263,300 3.51 2.8 Notes: Discount rate = 3%; cost factor F = 10. Table 5-6. Cost-effectiveness scores. The initial cost of the improvement is estimated based on the labor, equipment, and materi- als required to complete the work. In addition to these costs, the agency will have internal costs. These include, at a minimum, internal staff labor associated with managing, overseeing, and inspecting the work. The staffing section in Chapter 7 provides a method of estimating agency oversight costs. To capture the costs over the life of the asset to facilitate comparisons of the alternatives, the annual costs and future costs must be included. Table 5-5 highlights the annual energy savings associated with the ventilation upgrade and new LED lights as well as the savings in maintenance costs if the ceiling is removed permanently. Once the costs are estimated, the present value of all costs can be calculated using the present worth formulas in Section 5.3.1.1. It should be noted that the present value in the example is calculated using the service life after implementing the preservation action; to compare LCC directly at this point, the present value would need to be calculated based on a common time period, such as 100 years. However, the present values in Table 5-5 are used to calculate the annualized costs in Table 5-6. 5.3.2 ADT—The Effect of the Number of Users When an agency’s tunnels are located both in urban areas with high traffic volumes and in rural areas with significantly less traffic, prioritizing preservation actions is complicated. The risk of not implementing a preservation action will be higher for the urban tunnel due to the greater number of users. For example, a problem with Tunnel 2, located in an urban area with ADT of 100,000 vehicles, would have a greater impact than a problem with Tunnel 1 with 40,000 vehicles per day in a rural area on an Interstate. The risk of safety-related incidents would be greater in

Measuring Effectiveness of Preservation Actions 31 Tunnel 2 for a given preservation action focused on improving safety if it is not implemented than for the lower-ADT Tunnel 1. The impact of having to close the tunnel would be greater for Tunnel 2 than Tunnel 1 due to the higher number of users. Therefore, the number of users affected by a preservation action should be taken into consideration when evaluating cost-effectiveness. ADT data, which are typically collected by most agencies for their tunnels, afford a convenient measure for the number of users. As previously discussed, annualizing the LCC allows the user to compare actions with varied impacts on asset remaining life. Using the LCC and the ADT for the tunnel, an estimate of cost per user can be calculated and used to assess the cost-effectiveness of the preservation action. 5.3.3 Calculating the Cost-Effectiveness Score Improvements with the lowest cost per user are considered to be most cost-effective. To cal- culate the cost per user, LCC must be assessed and annualized. Therefore, for each preservation action, the following items are needed to calculate the CE score: • Capital cost (the initial cost of the preservation action in present-value dollars; includes labor and equipment); • Agency oversight cost (generally taken as a percentage of the capital cost)—used to add agency costs to overall project cost; • Change in annual costs considering energy, maintenance, closures, reduction in accidents, reduction in staff, and so forth; • ADT; and • Service life after improvement—the number of years to which the annualized cost applies. The annual life-cycle cost (ALCC) is computed as follows: ALCC 1 1 1 Equation 5-5C i i i An n( ) ( )= ∗ ∗ +  + −  − where ALCC = annual life-cycle cost ($ per year), C = capital cost + agency oversight cost ($), i = discount rate (%), n = change in service life resulting from improvement (years), and A = annual change in costs ($; costs associated with energy, maintenance, closures, reduc- tion in accidents, reduction of staff, etc.; negative if cost savings). The cost-effectiveness of the preservation action is inversely proportional to the cost per user. Accordingly, the reciprocal of the cost per user is used as an initial step in determining the CE score (yielding users per net dollar invested). Due to the range of possible costs for preservation actions and ADTs for tunnels, the resulting score has significant variability. To facilitate combination with the other scores to obtain an overall MOE for each preservation action, the CE score must be between 0 and 100. However, for low-cost improvements, particularly those in high-traffic tun- nels, the resulting score can far exceed 100. For these cases, a maximum score of 100 is assigned. Through evaluation of many preservation action examples, the following equation was developed for the CE score: CE 100, if 100 ALCC ADT 100, otherwise Equation 5-6 CE 100 ALCC ADT F F [ ] [ ] ( ) ( ) = ∗ > = ∗

32 Guide for the Preservation of Highway Tunnel Systems where CE = cost-effectiveness score, ADT = average daily traffic, number of vehicles, ALCC/ADT = annual life-cycle cost per daily vehicle, and F = cost factor, varies (see Section 5.3.4). As shown in Table 5-6, the low cost of the repairs to the CO system and the active leak in the tunnel, when calculated directly, would exceed 100. Therefore, the CE score is limited to the maximum value of 100 for these improvements. 5.3.4 Calibrating the CE Score Because Equation 5-6 is normalized to result in CE scores that range between 0 and 100, it must be validated for typical preservation actions. This is essential to ensure practical scores that are distributed over a distinguishable range between 0 and 100 and that offer an accurate and fair assessment of cost-effectiveness. For the values provided in Table 5-6, a value of 10 was assigned as the cost factor F in Equation 5-6. This number was selected based on the sample data for Agency X’s preservation actions and will require examination with any new set of data or significant changes to the records. Since the score is directly related to only the annual cost per daily vehicle, score equation calibration can be achieved through adjusting the cost factor variable to provide meaningful scores that: • Are distributed relatively evenly from 0 to 100, • Prevent too many repeat scores of 0 or 100, and • Properly reward actions that have low capital costs and affect a large number of vehicles. The cost factor variable F represents a scaling factor and allows the agency to adjust the CE score equation to best fit its preservation action costs. For example, an agency with a high number of expensive actions may determine that a cost factor equal to 20 best distributes its actions. An agency with less expensive actions may determine that a cost factor equal to 2 better fits its data. To determine the appropriate factor, it is recommended that the agency start with a cost factor of 10. Then, comparing all of its preservation actions together, if distribution is poor (within the range from 0 to 100) and too many actions are receiving a CE score of 100, the agency should incrementally increase its cost factor. If the distribution is poor and too many actions are receiving extremely low CE scores (near 0), then the agency should incrementally decrease its cost factor accordingly. Agency X chose to use a cost factor of 10 based on all of its preservation actions (presented in Appendix D). Note that while Table 5-5 shows a poor distribution of scores, these six preservation actions represent only a portion of Agency X’s evaluated actions. While an agency can theoretically use any cost factor in the CE core equation, it is important that the factor’s influence on all of the preservation actions’ CE scores be assessed. Arbitrarily choosing a factor without evaluating the results can lead to a poor distribution of CE scores and, therefore, an undesired impact on the overall MOE score (Section 5.5). The overall impact on the MOE score is dependent on the weight placed on the CE score. An agency placing a greater weight on cost must carefully evaluate the CE score and cost factor. 5.4 Implementing the Most Urgent Improvements When evaluating tunnel preservation priorities, the most urgent improvements should receive higher priority. Risk is an essential factor in determining these priorities. The risk associated with doing nothing (i.e., not implementing the preservation action) could result in an unsafe

Measuring Effectiveness of Preservation Actions 33 condition or a condition requiring closure of the tunnel. There are several factors that contribute to urgency: condition, remaining life, regulatory requirements, and unplanned events. Typically, elements in poor condition or near the end of their useful life have the greatest urgency for improvement due to the risk that exists if the improvement is not implemented. Upgrading to meet a new regulatory requirement or design standard may also be a priority for the agency. Furthermore, improvements focused on reducing risks associated with unplanned events such as floods or fires may be considered a high priority depending on the probability of the event. The RBU score, the third component contributing to the overall MOE, is determined by considering all of these factors. 5.4.1 Remaining Life Versus Service Life When an asset is near the end of its service life, its remaining life may be evaluated to determine whether replacement is warranted. Remaining life is dependent on multiple factors, including age, condition, and maintenance history. The RBU score evaluates condition as a separate factor (Section 5.4.2), but the age of the asset and how close it is to the end of its service life must be closely examined in evaluating how urgently the preservation action is needed. Assets that have expended most of their useful service lives should receive consideration of higher priority in the rankings based on the risks associated with their failure. For evaluation of the asset’s age and to facilitate comparison of multiple assets with varying service lives, it is recommended that the percent of life expended be determined. For each preser- vation action, the existing asset should be evaluated prior to implementation of the preservation action since the urgency applies to the existing asset. The remaining life for the existing asset must first be determined. There are considerable documentation and available references for obtaining service life information and calculating remaining life. While agencies may have significant experience with roadway pavement and structural repairs, it is important to accurately assess the service life of electrical and mechanical assets. The service lives for electrical and mechanical assets are highly dependent on maintenance performed, operating ambient temperature, and other environmen- tal conditions such as the presence of moisture and dust.(17) The average expected service life for most major electrical equipment can be assumed to be approximately 40 years. Appropriate adjustments can then be made to account for performed maintenance and operating conditions. The American Society of Heating, Refrigerating, and Air-Conditioning Engineers’ (ASHRAE) 2011 ASHRAE Handbook—Heating, Ventilating, and Air-Conditioning Applications details service life estimates in Chapter 37.(18) For tunnel systems, service life estimates are provided for centrifu- gal and axial fans, base-mounted pumps, motors, motor starters, transformers, and pneumatic, electric, and electronic controls. These values may also be assessed based on previous experience, historical data, or manufacturer’s recommendations. Once the remaining life and the service life are determined, the percent of life expended is eas- ily calculated. Table 5-7 presents the calculation for the six example preservation actions. When a preservation action implements a system that did not exist in the tunnel previously, “N/A” is entered, which eliminates age as a factor in the determination of the RBU score. 5.4.2 Determining Condition Tunnel conditions can be assessed by the tunnel operations and maintenance staff, by agency engineers, or by consultants engaged to perform in-depth condition surveys of the tunnel and its associated systems. Conditions are evaluated by any of these sources or in combination with one another to provide an assessment of the existing tunnel asset prior to developing the list of

34 Guide for the Preservation of Highway Tunnel Systems needed improvements. Such conditions can be established by following the guidelines of the FHWA 2015 Specifications for National Tunnel Inventory. Condition states of the existing asset or tunnel system, without consideration of any improve- ment, should be categorized as CS1 to CS4 as follows: • CS1—good condition, • CS2—fair condition, • CS3—poor condition, and • CS4—severe condition. This condition data should be determined at the time the list of preservation actions is devel- oped, making the data readily available for use in calculating the RBU score. As shown in Table 5-7, CS1 to CS4 represents a condition rating of 1 through 4, respectively. 5.4.3 Regulatory Compliance The majority of tunnels in existence today were designed before the introduction of NFPA 502 and its fire and life safety requirements. For this reason, tunnel owners are not required to upgrade to meet the standard, but many owners are upgrading to meet this standard where possible. Some owners are investigating the upgrades needed to achieve compliance for safety reasons, so these improvements may be given higher priority. Other codes or standards may also be applicable; thus, owners may be required to upgrade to meet current codes. It is therefore important to note preservation actions that are related to regulatory compliance. As with condition data, an indication of whether the preservation action is based on a regulatory requirement should be captured when the list of preservation actions is developed. A yes (Y) or no (N) indicator is provided, as shown in Table 5-7. 5.4.4 Risk of Unplanned Events The aftermath of Hurricane Sandy along the East Coast is evidence of the devastation that can result from unplanned events. Extreme weather, flooding, fires, and seismic events are among the many unplanned events that might occur, albeit rarely. Tunnel owners should consider the Table 5-7. Risk of unplanned events probability. Preservation Action Tu nn el # R em ai ni ng L ife Th eo re tic al S er v ic e Li fe % L ife E xp en de d C on di tio n (1 to 4) R eg ul at or y C om pl ia nc e I ss ue ? R isk o f U np la nn ed Ev en t P ro ba bi lit y (1 to 3) Ventilation upgrade to NFPA 502 1 1 25 96 2 Y 3 Install new LED lights 1 5 20 75 3 Y 1 CO system – repair to operating condition 2 2 20 90 4 Y 1 Repair active leak in ceiling 4 5 50 90 3 N 1 Remove existing concrete tunnel ceiling 6 0 50 100 3 N 1 Install flood gates 6 N/A* 100 N/A N/A N 3 *The condition is rated as N/A when the proposed preservation action is installing a component that is new to the tunnel system.

Measuring Effectiveness of Preservation Actions 35 risk of these events and their possible impacts and plan for these occurrences. Preservation actions focused on resiliency may be difficult to prioritize because the future is unknown, but owners must evaluate the risks and consider these risks as another factor when developing the RBU score. For each preservation action, the risk of potential unplanned events should be assessed and assigned a value of 1 to 3, with 1 representing a low probability and 3 representing a high prob- ability of the event occurring. 5.4.5 Establishing the RBU Ratings and Score As indicated, the urgency of a preservation action requires careful consideration of the many factors noted previously. For each of these factors, the associated risks of not implementing the improvement need to be contemplated. There is no simple formula to combine all of these factors; each tunnel may have different risks associated with it, based on location, physical geography, original construction, age, and so forth. Therefore, a means of considering these factors to obtain the RBU ratings and score is needed. A subjective approach is warranted where the various con- tributing factors are assessed, and a rating of the urgency of the improvement is assigned based on a scale of 0 to 10, where 0 represents a condition of like new with no risk, and 10 represents a critical or urgent condition (i.e., highest priority). This assessment should be performed by the AAMT, with the specific knowledge and understanding of the various systems and the risks associated with the tunnels. While the score is determined subjectively, it must be arrived at by logically looking through the RBU considerations, including percent life expended, condition, regulatory compliance, and risk of unplanned events. Once these considerations are established, each preservation action’s associated urgency can sensibly be determined (Table 5-8) as nonexistent (RBU rating of 0), low (RBU rating of 1 to 3), medium (RBU rating of 4 to 6), high (RBU rating of 7 to 9), or extreme (RBU rating of 10). Once categorized, the rating can be assigned at the bottom, middle, or top of the set range. Table 5-8 shows how this subjective evaluation may proceed, considering the various factors that must be considered in determining the RBU ratings. Because one of the risk factors may represent a perceived major risk and be far more significant than the other factors, it is not appropriate to merely place higher emphasis based on the number of applicable risk factors. To dismiss one high rating and assign a low overall RBU rating may ignore a significant and highly probable risk. However, minor risks for multiple factors included in the evaluation may sum to a high RBU rating. For this reason, the AAMT must evaluate all the risks and consider their relative significance in assigning an RBU rating. Table 5-8. Assigning RBU ratings. Risk Factors Urgency RBU Rating At least one area suggests the need for immediate action. Extreme 10 Multiple areas of consideration are of concern, or one area of concern is highly probable and would have significant impact on the LOS. High 9 8 7 At least one area of consideration is of concern. Medium 6 5 4 No areas of consideration are considered critical. Low 3 2 1 No indication of urgency. Nonexistent 0

36 Guide for the Preservation of Highway Tunnel Systems A selection of Agency X’s RBU ratings is provided in Table 5-9; additional preservation actions are presented in Appendix D. The rationale for assigning each of the RBU ratings is explained in the following. • The highest assignment made, a rating of 10, was for removal of the existing ceiling in Tun- nel 6 since it is in poor condition and has exceeded its useful life. It represents a significant risk due to the deterioration of the ceiling and is a safety hazard to the public traveling beneath it. • The ventilation upgrade in Tunnel 1 was assigned an RBU rating of 8. While the existing ventilation system is in fair condition, it requires continuous maintenance, the equipment is getting very old, and parts are becoming difficult to find. Agency X wants to upgrade it to meet NFPA 502 to improve ventilation during a fire event, making this a regulatory compliance issue. • Both the CO system (Tunnel 2) and active leak (Tunnel 4) repairs were assigned an RBU rating of 7 due to the condition and age of these assets. Since the CO system is currently non-operational, it received a severe condition rating. Tunnel 4 is in poor condition due to active leaks that represent a significant safety risk, especially during months when freezing occurs. • There is a risk of flooding in Tunnel 6, and there is currently no flood protection system in place. However, a rating of 6 has been assigned to this action since the owner deemed the probability and ultimate impact of a flood event to be noncritical. • Installing new LED lights in Tunnel 1 would help the owner reduce maintenance even though the existing lights have a few years of service life left and are in relatively good condition. Lights are required in tunnels, so there is a regulatory requirement, but the existing lighting would also meet the requirement. Based on the risk assessment, it is assigned an RBU rating of 3. Once a rating of 0 to 10 is assigned, the RBU score is calculated by multiplying the rating by 10 to achieve a score of between 0 and 100 (Table 5-9). Pr es er va tio n A ct io n Tu nn el # R em ai ni ng L ife Th eo re tic al S er v ic e Li fe % L ife E xp en de d C on di tio n (1 to 4) R eg ul at or y C om pl ia nc e I ss ue ? R isk o f U np la nn ed Ev en t P ro ba bi lit y (1 to 3) R BU R at in g (1 to 10 ) R BU S co re Ventilation upgrade to NFPA 502 1 1 25 96 2 Y 3 8 80.0 Install new LED lights 1 5 20 75 3 Y 1 3 30.0 CO system – repair to operating condition 2 2 20 90 4 Y 1 7 70.0 Repair active leak in ceiling 4 5 50 90 3 N 1 7 70.0 Remove existing concrete tunnel ceiling 6 0 50 100 3 N 1 10 100.0 Install flood gates 6 N/A 100 N/A N/A N 3 6 60.0 Table 5-9. Risk-based urgency score.

Measuring Effectiveness of Preservation Actions 37 5.4.6 Calibrating the Risk-Based Urgency Score There is considerable subjectivity in the determination of the RBU score. In addition, the scores are further aggregated by the multiplication factor of 10. Therefore, the AAMT’s determination of the urgency might appear to have a significant impact on the final prioritization. The final RBU score is calculated based on the subjectively determined RBU rating assigned a value of between 0 and 10. As outlined previously, the RBU rating must be arrived at by considering the RBU criteria, including percent life expended, condition, regulatory compliance, and risk of unplanned events. Once the criteria have all been addressed, each preservation action’s associated urgency can be evaluated using Table 5-8 to first consider the urgency as nonexistent (RBU rating of 0), low (RBU rating of 1 to 3), medium (RBU rating of 4 to 6), high (RBU rating of 7 to 9), or extreme (RBU rating of 10). Once categorized, the rating can be assigned at the bottom, middle, or top of the set range. As long as the preservation action is categorized with this method, the most the RBU rating can vary is two points. While this multiplies to 20 points in the RBU score, the RBU score is one of three scores used to determine the overall MOE score (Section 5.5). Still, to maintain the method’s integrity, an agency assigning a high weight to the RBU score must carefully and method- ically evaluate each preservation action using a systematic approach due to the subjective nature of this score. Accurate asset life and condition data are essential for appropriate consideration. The example presented in Table 5-10 details the RBU rating assigned by a second reviewer at Agency X. The second reviewer considered all of the same information as the first reviewer (Table 5-9) and came up with some variations in RBU ratings. Analysis of the second reviewer’s choices: • Ventilation upgrade to NFPA 502: Classified as high risk, the second reviewer chose to rate this action as a 9, feeling that the little remaining asset life and high probability of an unplanned event (fire) justified a higher rating than the 8 assigned by the first reviewer. The difference of 1 results in a 10% difference in RBU score and ultimately a 4.5-point difference when Pr es er va tio n A ct io n Tu nn el # R em ai ni ng L ife Th eo re tic al S er vi ce Li fe % L ife E xp en de d C on di tio n (1 to 4) R eg ul at or y C om pl ia nc e I ss ue ? R isk o f U np la nn ed Ev en t P ro ba bi lit y (1 to 3) R BU R at in g (1 to 10 ) R ev iew er 1 R BU R at in g (1 to 10 ) R ev iew er 2 Ventilation upgrade to NFPA 502 1 1 25 96 2 Y 3 8 9 Install new LED lights 1 5 20 75 3 Y 1 3 3 CO system – repair to operating condition 2 2 20 90 4 Y 1 7 9 Repair active leak in ceiling 4 5 50 90 3 N 1 7 8 Remove existing concrete tunnel ceiling 6 0 50 100 3 N 1 10 10 Install flood gates 6 N/A 100 N/A N/A N 3 6 6 Table 5-10. Risk-based urgency rating—second reviewer.

38 Guide for the Preservation of Highway Tunnel Systems considering the final measure of effectiveness score (Section 5.5; RBU score assigned a weight of 45% of total combined score). • Install new LED lights: The second reviewer agreed with choosing the higher end of the low risk rating range and assigned the same rating as the first reviewer. • CO system—repair to operating condition: Both reviewers categorized the CO system as a high risk. However, the second reviewer felt it should be rated at the top of the high-risk category, and the first reviewer assigned a rating at the bottom of the category. This rating difference of 2 results in a 20% difference in the RBU score and a 9-point difference in the RBU score component of the overall MOE. • Repair active leak in ceiling: Classified as a high-risk item. Reviewer 1 felt that the repair should be rated at the bottom of the category, and Reviewer 2 rated it at the middle of the category, resulting in an RBU rating difference of 1 point. • Remove existing concrete tunnel ceiling: The concrete tunnel ceiling is currently an extreme risk, and therefore both reviewers assigned an RBU rating of 10. • Install flood gates: Overall, this action is classified as medium risk. However, due to the risk of Tunnel 6 flooding, both reviewers felt the top category rating of 6 should be assigned. As discussed, while the assigned RBU ratings are subjective, using a logical process starting with the risk categories described, an AAMT can effectively develop the RBU score. 5.5 Measure of Effectiveness The metric presented in this chapter uses an overall MOE as the indicator in establishing the priority of preservation actions to be implemented. The MOE for a proposed preservation action is calculated by combining the LOS score, the CE score, and the RBU score. It is based on a scale of 0 to 100 and provides a rational means to prioritize a diverse list of preservation actions. The three scores can be weighted differently based on the agency’s particular situation or overall goals. This is another opportunity to customize the metric for a specific agency and its tunnel assets. The AAMT should establish the weights to be applied to the three individual scores in consideration of its tunnels and overall agency goals and objectives. For the example provided in Table 5-11, the AAMT weighted the three scores as follows: agency LOS = 35%, cost- effectiveness = 20% and risk-based urgency = 45%. Levels of Service Level of Service Score Cost- Effectiveness Score Risk-Based Urgency Score MOE Score (Eq. 5-7)Weights 35% 20% 45% Preservation Action Tunnel# Ventilation upgrade to meet NFPA 502 1 66.0 1.9 80.0 59.5 Install new LED lights 1 78.0 2.4 30.0 41.3 CO system – repair to operating condition 2 63.2 100.0 70.0 73.6 Repair active leak in ceiling 4 89.0 100.0 70.0 82.7 Remove existing concrete tunnel ceiling 6 85.4 2.3 100.0 75.4 Install flood gates 6 66.0 2.8 60.0 50.7 Table 5-11. Measure of effectiveness score.

Measuring Effectiveness of Preservation Actions 39 Measure of effectiveness is computed as follows: MOE score LOS CE RBU Equation 5-7LOS CE RBUW W W= ∗ + ∗ + ∗ where LOS = level-of-service score, CE = cost-effectiveness score, RBU = risk-based urgency score, and WLOS, WCE, WRBU = weights for the LOS, CE, and RBU, where (WLOS + WCE + WRBU) = 100. In Table 5-11, the highest MOE is for the repair of the active leak in the tunnel ceiling. This preservation action had relatively high values for all three scores, although the most heavily weighted RBU score was not the highest of the improvements being compared. The next high- est scoring preservation action received the maximum RBU score, a significantly high impact on Agency X’s LOS, but a low CE score. While the CO system repair received the maximum CE score, it ranked third in the list due to its moderate impact on the agency’s LOS and lower RBU score. Observation of the data in this manner shows that the resulting MOE score is not easily predicted since the weighting performed in this final step affects the final outcome.